Skip to content
From Fire Safety Academy Wiki
Revision as of 18:52, 2 November 2026 by Mark (talk | contribs) (Revise against the FSE wiki writing guidelines: verified and corrected citations, added sources, See also, links, style (revised by Claude Opus 5.5))
Contents
  1. 1How thermal runaway starts
  2. 2Gases, chain reactions and explosions
  3. 3Heat release and fire development
  4. 4Toxic and corrosive products
  5. 5Extinguishing and cooling
  6. 6Firefighting and salvage in car parks
  7. 7Prevention in car parks
  8. 8Uncertainties and limits
  9. 9See also
  10. 10Notes
  11. 11References

A lithium-ion battery fire is a fire in a lithium-ion battery or battery pack that is driven by thermal runaway. Thermal runaway is a failure mechanism in which a battery cell heats itself up, and it can result in fire.[1] Many electric cars store their energy in lithium-ion packs, which can hold 100 kWh or more.[2] Once thermal runaway has started, heat passes from cell to cell, flammable gases are released and the fire keeps re-igniting until the energy in the pack is used up. The pack is sealed against moisture and dirt, so water cannot reach the cells directly and cooling needs a lot of water for a long time.[1]

In tests, burning electric cars reached temperatures and peak heat release rates similar to those of petrol and diesel cars.[3][4] Their fires nevertheless develop more slowly, last longer, can involve repeated small explosions and flash fires, and release more hydrogen fluoride, a toxic and corrosive gas.[5][6] Together with high voltages, these risks are one reason why firefighting in indoor car parks may be limited or impossible.[7] A 2021 study by the Netherlands Institute for Public Safety (NIPV) on indoor car parks with electric vehicles found that the probability of such fires could hardly be determined from empirical data.[8]

How thermal runaway starts

A battery pack in an electric car consists of cells of about 4 V, joined into modules of up to 60 V that together form a pack of 300 to 1,000 V.[2] The SFPE Handbook explains the general principle of self-heating: heat produced by a chemical reaction competes with heat lost to the surroundings. Most reaction rates rise almost exponentially with temperature, but heat losses by conduction rise only linearly. Ignition, or thermal runaway, therefore begins where the temperature is highest, if it occurs at all.[9]

In a battery, each pack has its own safety window, and beyond it the pack can heat up further and go into thermal runaway.[1] The temperature at which this can start depends on the type of battery: the NIPV report gives 70 to 250 °C in one place and 50 to 270 °C in another.[1][10] Every electric car has a battery management system that monitors the individual cells.[1] When limits are exceeded, it can switch on active cooling during charging, stop charging or disconnect the pack. The system is designed to be fail-safe, so a fault in the system itself also switches the pack off.[3]

A battery can fail because of a design or manufacturing error in the electrode, separator or electrolyte. This can lead to failures in the battery management system and to overcharging, but such errors are rare and, once detected, generally lead to a recall.[3] Outside causes include external heat, misuse and damage in a collision.[3] Parking and charging electric cars in car parks add causes that do not exist with petrol and diesel cars. Examples are a damaged cable or charging station, a fault in the battery management system, a collision with a charging station and a defective pack that has gone unnoticed.[5] Radiant heat from a neighbouring burning car may also cause thermal runaway in an electric car parked next to it.[11]

How often batteries fail is uncertain. Estimates quoted in the NIPV report range from one failure in a million or fewer to about one in 100 million. The report concludes that the probability of battery failure seems relatively low. Still, no lithium-ion battery is intrinsically safe: no pack design can completely prevent the release of enough energy for thermal runaway.[3]

Gases, chain reactions and explosions

A heating cell can release flammable gases. These may ignite if there is enough oxygen and the ignition temperature is reached, and the casing of the cell may be hot enough for this. The heat of one cell can also push the cells next to it into thermal runaway, so that the failure spreads through the pack as a chain reaction.[1] Gases held in a pouch or cartridge can escape if the pack or its pressure valve fails under high pressure. If the pack is mechanically damaged, the pressure wave can throw out parts and fragments, which may be burning.[1]

Gas that does not ignite at once can build up into an explosive mixture in a confined or poorly ventilated space.[12] If the mixture reaches its lower explosion limit, the lowest concentration at which a flame can spread through it, and is then ignited, a gas explosion can occur.[12][13] Among the gases that burning packs give off are hydrogen, methane and carbon monoxide.[14] Hydrogen alone has a lower flammability limit in air of 4.1 per cent by volume, and its flame is often invisible in daylight.[13] The NIPV report counts repeated small explosions, flying parts and flash fires among the ways in which fires in electric cars differ from other car fires.[5]

One hot cell heats neighbours, releasing gases that accumulate and can ignite or explode.
Thermal runaway can propagate and produce flammable gases. Based on Van der Graaf et al. (2021/2023), p. 22 and p. 27.

Heat release and fire development

In terms of temperature, fires in electric cars do not differ significantly from those in conventional cars. Battery tests reviewed in the NIPV report measured maximum flame temperatures of 798 to 863 °C, rising with how fully the batteries were charged. Full-scale tests on both types of car measured 800 ± 100 °C.[3] The heat radiation from an electric car with a fully charged pack was also similar to that from a conventional car.[4]

The heat release rate, the amount of heat a fire releases per second, largely determines how intense a fire is.[15] A review of published tests found similar peaks of about 6 MW for electric and fossil-fuelled cars. The total heat released was 4.7 to 8.5 GJ for electric cars and 3.3 to 10 GJ for the others.[4] Larger batteries raise the peak: packs of about 15 to 25 kWh gave about 4 to 6 MW, while packs of more than 100 kWh give about 8 MW. That is above the 6.8 MW of an average mid-range conventional car.[4] Introduction to Fire Safety Engineering gives a peak of 8.5 MW for a car and 2.5 MW for an average sofa.[15]

The course of the fire over time differs more. In a petrol or diesel car, the plastic tank can fail and suddenly release fuel that catches fire, so the temperature rises very rapidly in the first minutes. Battery cells first need time to warm up and give off gas. The temperature therefore rises evenly, at about 0.2 °C per minute, until thermal runaway makes it climb by more than 10 °C per minute.[14] A normal passenger car fire generally burns for just over half an hour and peaks at about 5 MW after 10 to 15 minutes. An electric car left to burn without intervention burns out in about 90 minutes, and an extinguished one can keep re-igniting for several hours.[14]

An electric vehicle fire in a car park can spread from car to car as a travelling car fire. The NIPV report expects electric cars to make such fires more likely, because a long-burning pack radiates heat to its neighbours for longer.[5] In 2021, Dutch building regulations, standards and guidelines were still based on fires in conventionally fuelled cars, and modern cars contain more plastic than older steel models.[7] The car-fire curve in the Dutch standard NEN 6098 rests on a design fire set by TNO in 1999. In the TNO tests, fire could spread to the adjacent cars 10 to 15 minutes after the first one. With more plastics in modern cars, such curves may be less conservative in some situations.[16]

Toxic and corrosive products

The toxic hazard of a fire, the subject of fire toxicity, depends on how quickly and how much toxic products are produced, and on how potent they are.[17] Burning battery packs release corrosive substances from the electrolyte and combustion products from the plastics and packaging. Besides methane, carbon monoxide, carbon dioxide and hydrogen, they release the toxic gases hydrogen fluoride and hydrogen chloride.[14] The SFPE Handbook counts these acid gases among the most important irritants in fire smoke. Irritants can impair escape or incapacitate people through pain and breathing difficulty.[17]

A study cited in the NIPV report found about 20 to 200 mg of hydrogen fluoride per watt-hour of battery capacity. A 100 kWh pack can thus emit 2 to 20 kg, depending on its charge. Comparable tests on electric and conventional cars found hardly any difference in other toxic gases, but the electric cars appeared to produce about 1.8 times more hydrogen fluoride.[14] Hydrogen fluoride binds to other substances and smoke particles and settles on floors, walls and ceilings. In one study this more than halved its concentration in the smoke within 10 to 20 minutes, but made firefighting clothing more contaminated.[14] The report requires the clothing to be decontaminated because of the toxic and corrosive gases, unlike after a conventional car fire.[6]

Battery smoke travels through air and deposits on clothing, walls, floors and residue.
Battery-fire contamination includes airborne and deposited hazards. Based on Van der Graaf et al. (2021/2023), pp. 26, 29–30, 39 and 47.

For people escaping, the extra hydrogen fluoride changes little. The plastic parts of any burning car already create severely toxic smoke, and the report puts the effect of the extra hydrogen fluoride on escape safety at zero. Systemic health effects seem unlikely, although slight exposure may irritate the skin and severe exposure may cause chemical burns.[6] Research cited in the report found that firefighters in protective equipment can attack safely with proper hygiene if their exposure to the smoke lasts no more than 20 to 30 minutes.[14] The report sees no reason to assume that they need more protective equipment than at a fire in a conventional car.[18]

Smoke with extra hydrogen fluoride can spread from a car park through open doors into the escape routes and stairwells of buildings above.[6] People who leave at once are exposed only briefly, but those who cannot escape or take longer are exposed for longer. For this reason the report advises against placing charging stations near pedestrian escape routes.[19]

Extinguishing and cooling

Fighting a battery fire involves two tasks: putting out the flames, and cooling the pack for a long time so that thermal runaway does not set it alight again. Putting out the flames needs no more capacity than for a conventional car.[20] Stopping thermal runaway is much harder, because the closed construction of the pack keeps water away from the cells.[1] According to the NIPV report, a burning pack can only be controlled by prolonged cooling with a lot of water, and extinguishing it requires water inside the pack.[6]

After the flames are out, the pack catches fire again within a short time. It therefore needs continuous cooling, and its temperature has to be monitored for a long time.[20] Firefighters can extinguish continuously or each time the fire re-ignites, and the amount of water needed depends on the pack and how fully it is charged.[20] For packs without special facilities, the most successful way to get water inside is to immerse the pack in water.[6] An immersion container is currently the best method available for long-term cooling, but its effect depends on how far the water can reach the cells.[20]

Flames are knocked down before prolonged cooling, temperature monitoring and salvage.
Battery-fire response continues after visible flames disappear. Based on Van der Graaf et al. (2021/2023), pp. 45–47.

Fixed sprinkler or water mist systems cannot extinguish a fire in an electric car, because the water droplets cannot enter the pack. They can, however, reduce the probability that the fire grows and spreads to nearby structures and vehicles.[21] Covering the car in foam is not a useful method for the time being. The fire can re-ignite under the foam, because packs in thermal runaway can re-ignite even when little oxygen is available.[18]

Fire classes sort fires by the type of fuel. In the US scheme of NFPA 10 (2002 edition), summarised in NUREG-1805, fires in energised electrical equipment form Class C, for which the agent must not conduct electricity. Fires in combustible metals, among them lithium, form Class D, on which water-based extinguishers should never be used.[22] The NIPV report, which deals with lithium-ion packs in cars, does not use these classes and relies on water.[6] Although electric cars carry up to about 400 V, with peaks of up to 800 V, the probability of electrocution is negligible when the fire is fought from a distance.[7][23]

Firefighting and salvage in car parks

In an indoor car park, a fire in an electric car adds several problems for the fire service. It is uncertain how the fire will develop and spread, and the pack can re-ignite and burn for longer. Explosions, flare fires and specific toxic gases can occur, and the deployment may be long and need a lot of water.[23] The stability of the structure is uncertain, and a burning electric car may even start to move.[6] The report concludes that the risks and uncertainties of an attack have increased compared with car parks without electric cars.[23]

Dutch fire services prefer to fight a fire from the outside. Before choosing an attack, they establish whether the location of the fire is known, whether it can be reached from outside and whether enough cooling capacity is available. Only if all three conditions are met can the fire be put out from the outside. Otherwise the attack becomes defensive, and a large structure such as a car park is in principle considered beyond saving.[24] For car parks, the report concludes that an attack from inside is generally not safely possible, although the commanding officer may still decide on one.[25] The fire service will not automatically enter, so a car park may burn out.[18]

Cooling capacity is estimated with rules of thumb. A high-pressure jet of 125 litres per minute is generally sufficient for a heat release rate of up to 2.5 MW. A low-pressure jet of 450 litres per minute generally suffices for up to 10 MW.[25] Fires in electric cars call for low-pressure jets, and several burning cars will probably need several jets.[24] Introduction to Fire Safety Engineering explains the idea behind such rules: the expected heat release rate of a fire indicates the cooling capacity needed to fight it.[16]

Because a pack can re-ignite after the fire is out, the report advises planning how an electric car can be removed from the car park. Together with a properly equipped salvage company, the fire service removes the car, which can then be left to react in a water tank until thermal runaway has stopped.[26] Not every salvage company has the right equipment, and smoke or poor access can delay removal from an underground car park.[20] In a fire in the Singelgarage car park in Alkmaar in 2020, the pack of one electric car kept re-igniting. A police robot towed the car out, and it was then cooled for a long time in an immersion container.[27]

Ordinary extinguishing water from electric car fires harms the environment hardly more than that from other car fires and can go into the sewer.[20] Cooling water in an immersion tank is different. Burnt packs contain lithium oxide, which reacts with water to form lithium hydroxide, so the pH of the water can keep rising.[5] Such cooling water may not be discharged directly into the sewer, and a residue with a neutral pH on the outside may still contain heavy metals.[18]

Prevention in car parks

In the Netherlands, the 2021 NIPV report discusses prevention against the Building Decree 2012, whose fire safety rules did not specifically address parking and charging electric cars.[28] Under that decree, enclosed indoor car parks of more than 1,000 m² needed an automatic fire alarm system with full detection, and new indoor car parks had to have charging facilities.[21][29] Amendments proposed in 2021 for its successor, the Building Works and Living Environment Decree, would make sprinklers mandatory in certain car parks. They would also require a facility to switch off all charging points at once.[10]

The report offers its measures as a range from which to build a tailor-made set, not as a detailed standard.[30] It sees a sprinkler or water mist system as the most obvious way to hold a fire to a few cars and so prevent a travelling fire.[31][32] Fast detection helps, but its effect depends on prompt follow-up of the alarm.[21][33] Carbon monoxide detectors gave the fastest alert in one study, but in the Netherlands they are not used as fire detection.[21] Collision protection for charging stations is a further measure.[31]

The report prefers parking spaces for electric cars near the vehicle entrances and at street level, so that a fire can be fought from outside and a car removed quickly.[34] It advises against places where ventilation air is supplied.[19] Grouping charging points in one marked area helps the fire service find electric cars, but such clustering can also lead to a travelling car fire.[26] Users can be told not to park a car whose battery management system reports a fault and to keep their distance if gases are released.[33]

Uncertainties and limits

How often electric cars catch fire is largely unknown, because there are no reliable datasets.[35] The report judges that a parked electric car is unlikely to catch fire more often than a fossil-fuelled car. Charging, however, is expected to raise the probability, because it adds an activity, the technology is still in its teething phase and existing electrical installations may not be safe enough.[36] Ageing probably raises the internal resistance of the cells, which increases the heat they produce. An older battery also holds a lower maximum charge, however, so its thermal runaway is less intense and releases fewer flammable gases.[36]

The NIPV report covers only lithium-ion passenger cars in Dutch indoor car parks, and "lithium-ion" itself covers many kinds of lithium batteries that respond differently to fire.[10] It warns that test results vary greatly with the amount of fuel, the battery capacity and the plastics in the car, so single studies cannot simply be generalised.[8] Battery packs can also be combined into systems of 10 to 400 kWh, for example in trucks and electric ships. Stationary installations such as a battery energy storage system lie outside the report's scope.[2][10] Open questions include the exact process of thermal runaway, the conditions under which packs re-ignite and the best detection and suppression systems.[37]

See also

Notes

  1. ↑ 1.0 1.1 1.2 1.3 1.4 1.5 1.6 1.7 Van der Graaf et al. (2021/2023), p. 22.
  2. ↑ 2.0 2.1 2.2 Van der Graaf et al. (2021/2023), p. 21.
  3. ↑ 3.0 3.1 3.2 3.3 3.4 3.5 Van der Graaf et al. (2021/2023), p. 23.
  4. ↑ 4.0 4.1 4.2 4.3 Van der Graaf et al. (2021/2023), pp. 24–25.
  5. ↑ 5.0 5.1 5.2 5.3 5.4 Van der Graaf et al. (2021/2023), p. 29.
  6. ↑ 6.0 6.1 6.2 6.3 6.4 6.5 6.6 6.7 Van der Graaf et al. (2021/2023), p. 30.
  7. ↑ 7.0 7.1 7.2 Van der Graaf et al. (2021/2023), p. 11.
  8. ↑ 8.0 8.1 Van der Graaf et al. (2021/2023), p. 20.
  9. ↑ Gray (2002), pp. 2-211–2-212.
  10. ↑ 10.0 10.1 10.2 10.3 Van der Graaf et al. (2021/2023), p. 18.
  11. ↑ Van der Graaf et al. (2021/2023), pp. 32–33.
  12. ↑ 12.0 12.1 Van der Graaf et al. (2021/2023), p. 27.
  13. ↑ 13.0 13.1 Iqbal and Salley (2004), p. 16-2.
  14. ↑ 14.0 14.1 14.2 14.3 14.4 14.5 14.6 Van der Graaf et al. (2021/2023), p. 26.
  15. ↑ 15.0 15.1 Van Rossum (2027), §4.5.
  16. ↑ 16.0 16.1 Van Rossum (2027), §8.6.
  17. ↑ 17.0 17.1 Purser (2002), p. 2-91.
  18. ↑ 18.0 18.1 18.2 18.3 Van der Graaf et al. (2021/2023), p. 47.
  19. ↑ 19.0 19.1 Van der Graaf et al. (2021/2023), p. 39.
  20. ↑ 20.0 20.1 20.2 20.3 20.4 20.5 Van der Graaf et al. (2021/2023), p. 46.
  21. ↑ 21.0 21.1 21.2 21.3 Van der Graaf et al. (2021/2023), p. 37.
  22. ↑ Iqbal and Salley (2004), pp. B-13, B-16.
  23. ↑ 23.0 23.1 23.2 Van der Graaf et al. (2021/2023), p. 43.
  24. ↑ 24.0 24.1 Van der Graaf et al. (2021/2023), p. 44.
  25. ↑ 25.0 25.1 Van der Graaf et al. (2021/2023), p. 45.
  26. ↑ 26.0 26.1 Van der Graaf et al. (2021/2023), p. 42.
  27. ↑ Van der Graaf et al. (2021/2023), p. 12.
  28. ↑ Van der Graaf et al. (2021/2023), p. 17.
  29. ↑ Van der Graaf et al. (2021/2023), p. 10.
  30. ↑ Van der Graaf et al. (2021/2023), p. 34.
  31. ↑ 31.0 31.1 Van der Graaf et al. (2021/2023), p. 36.
  32. ↑ Van der Graaf et al. (2021/2023), pp. 49–50.
  33. ↑ 33.0 33.1 Van der Graaf et al. (2021/2023), p. 40.
  34. ↑ Van der Graaf et al. (2021/2023), pp. 41–42.
  35. ↑ Van der Graaf et al. (2021/2023), p. 54.
  36. ↑ 36.0 36.1 Van der Graaf et al. (2021/2023), p. 28.
  37. ↑ Van der Graaf et al. (2021/2023), p. 55.

References

  • Gray, B. (2002). Spontaneous combustion and self-heating. In P. J. DiNenno, D. Drysdale, C. L. Beyler, W. D. Walton, R. L. P. Custer, J. R. Hall Jr., & J. M. Watts Jr. (Eds.), SFPE handbook of fire protection engineering (3rd ed., pp. 2-211–2-228). National Fire Protection Association; Society of Fire Protection Engineers.
  • Iqbal, N., & Salley, M. H. (2004). Fire Dynamics Tools (FDTs): Quantitative fire hazard analysis methods for the U.S. Nuclear Regulatory Commission Fire Protection Inspection Program (NUREG-1805). U.S. Nuclear Regulatory Commission, Office of Nuclear Reactor Regulation.
  • Purser, D. A. (2002). Toxicity assessment of combustion products. In P. J. DiNenno, D. Drysdale, C. L. Beyler, W. D. Walton, R. L. P. Custer, J. R. Hall Jr., & J. M. Watts Jr. (Eds.), SFPE handbook of fire protection engineering (3rd ed., pp. 2-83–2-171). National Fire Protection Association; Society of Fire Protection Engineers.
  • Van der Graaf, J., Hessels, T., & Rosmuller, N. (2023). Fire safety of indoor car parks accommodating electrically powered vehicles. Nederlands Instituut Publieke Veiligheid. (Original work published 2021)
  • Van Rossum, E. (2027). Introduction to Fire Safety Engineering. Fire Safety Academy.